Neisseria meningitidis is a Gram-negative bacteria which colonizes the human upper respiratory tract and is responsible for worldwide sporadic and cyclical epidemic outbreaks of, most notably, meningitis and sepsis. The attack and morbidity rates are highest in children under 2 years of age.
Like other Gram negative bacteria, Neisseria meningitidis typically possess a cytoplasmic membrane, a peptidoglycan layer, an outer membrane which together with the capsular polysaccharide constitute the bacterial wall, and pili which project into the outside environment. These surface structures mediate infection and interact with the host immune system. For example, a first step in infection with Neisseria is adherence to target cells, which is thought to be mediated by the pili and, possibly, other adhesins such as Opc. Protein, phospholipid and polysaccharide components of the outer membrane have been reported to elicit an immune response.
Neisseria meningitidis spp. can be divided into serologic groups, types and subtypes on the basis of reactions with polyclonal (Frasch, C. E. and Chapman, 1973, J. Infect. Dis. 127: 149-154) or monoclonal antibodies (Hussein, A., MONOCLONAL ANTIBODIES AND N. MENINGITIDIS. Proefschrift. Utrecht, Nederland, 1988) that interact with different surface antigens. Serogrouping is based on immunologically detectable variations in the capsular polysaccharide. About 12 serogroups are known: A, B, C, X, Y, Z, 29-E, W-135, H, I, K and L (Ashton, F. E. et al., 1938, J. Clin. Microbiol. 17: 722-727; Branham, S. E., 1956, Can. J Microbiol. 2: 175-188; Evans, A. C., 1920, Lab. Bull. 1245: 43-87; Shao-Qing, et al., 1972, J. Biol. Stand. 9: 307-315; Slaterus, K. W., 1961, Ant. v. Leeuwenhoek, J. Microbiol. Serol. 29: 265-271). Currently, serogroup B (MenB) is responsible for about half to 80% of reported invasive Neisseria meningitidis diseases.
Serotyping is based on monoclonal antibody defined antigenic differences in an outer membrane protein called Porin B (PorB). Antibodies defining about 21 serotypes are currently known (Sacchi et al., 1998, Clin. Diag. Lab. Immunol. 5:348). Serosubtyping is based on antibody defined antigenic variations on an outer membrane protein called Porin A (PorA). Antibodies defining about 18 serosubtypes are currently known. Serosubtyping is especially important in Neisseria meningitidis strains where immunity may be serosubtype specific. Most variability between PorA proteins occurs in two (loops I and IV) of eight putative, surface exposed loops. The variable loops I and IV have been designated VR1 and VR2, respectively. Since more PorA VR1 and VR2 sequence variants exist that have not been defined by specific antibodies, an alternative nomenclature based on VR typing of amino acid sequence deduced from DNA sequencing has been proposed (Sacchi et al., 2000, J. Infect. Dis. 182:1169; see also the Multi Locus Sequence Typing web site). Lipopolysaccharides can also be used as typing antigens, giving rise to so-called immunotypes: L1, L2, etc.
Neisseria meningitidis also may be divided into clonal groups or subgroups, using various techniques that directly or indirectly characterize the bacterial genome. These techniques include multilocus enzyme electrophoresis (MLEE), based on electrophoretic mobility variation of an enzyme, which reflects the underlying polymorphisms at a particular genetic locus. By characterizing the variants of a number of such proteins, genetic “distance” between two strains can be inferred from the proportion of mismatches. Similarly, clonality between two isolates can be inferred if the two have identical patterns of electrophoretic variants at number of loci. More recently, multilocus sequence typing (MLST) has superseded MLEE as the method of choice for characterizing the microorganisms. Using MLST, the genetic distance between two isolates, or clonality is inferred from the proportion of mismatches in the DNA sequences of 11 housekeeping genes in Neisseria meningitidis strains (Maiden et al., 1998, Proc. Natl. Acad. Sci. USA 95:3140).
Given the prevalence and economic importance of invasive Neisseria meningitidis infections, it is not surprising that many attempts have been made to develop treatments. Although these infections can be treated with antibiotics, about 10 to 20% of treated patients die, and many survivors are left with permanent neurologic sequelae, such as amputation, neurosensory hearing loss, and paralysis. Also, microorganisms can develop antibiotic resistance. Thus, prevention with vaccines is a preferable mode to contain the spread of infection.
Because the polysaccharide capsule is one of the outermost structures of pathogenic Neisseria meningitidis, it has been a primary focus of attempts to develop vaccines. Different preparations of capsular polysaccharides have been used to control the outbreaks and epidemics of the serogroups A, C, Y and W-135, as mono-, di-, tri- or tetravalent vaccines (Gold et al., 1969-1970, Bull. WHO 45: 272-282; Gotschlich et al., 1969, J. Exp. Meal 129: 134-136; Hankins, 1982, Proc. Soc. Biol. Med. 169: 54-57; U.S. Pat. No. 6,080,589). However, capsular polysaccharide vaccines suffer from: poor or no-response to polysaccharide C in children under 2 years of age; thermolability of polysaccharide A; difficulties regarding the induction of immunologic tolerance after vaccination or re-vaccination with polysaccharide C (Granoff et al., 1998, J. Infect. Dis. 160: 5028-5030; MacDonald et al., 1998, JAMA 280:1685-1689; MacDonald et al., 2000, JAMA 283: 1826-1827). To circumvent these immunologic properties, polysaccharides from serogroups A and C have been covalently coupled to protein carriers to make “conjugate” vaccines. In contrast to plain polysaccharide vaccines, these conjugate vaccines are highly immunogenic in infants, upon re-injection elicit boostable increases in serum anticapsular antibody concentrations, and prime for the ability to generate memory antibody responses to a subsequent injection of plain polysaccharide (Campagne et al. 2000, Pediat. Infect. Dis. J. 19: 144-150; Maclennan et al., 2000, JAMA 283: 2795-2801). Conjugate vaccines with similar properties have been highly effective in preventing invasive diseases caused by other encapsulated bacteria, such as Haemophilus influenzae type b or Streptococcus pneumoniae. 
The capsular polysaccharide (PS) of serogroup B Neisseria meningitidis is a very poor immunogen in humans (Wyle et al., 1972, J. Infect. Dis. 126: 514-522; Zollinger, et al., 1979, J. Clin. Invest. 63: 836-834; Jennings et al., 1981, J. Immunol. 127: 104-108). Further attempts to improve the polysaccharide's immunogenicity through conjugation to protein have been unsuccessful (Jennings et al., 1981, J. Immunol. 127: 104-108). To enhance the immunogenicity, the meningococcal serogroup B capsule polysaccharide (MenB PS) has been chemically modified (N-propionylated group was substituted for the N-acetyl group of B polysaccharide) and coupled covalently to a protein carrier (N—Pr-MenB PS-protein) conjugate. The vaccine induces in mice high titers of IgG antibodies which are bactericidal and protective (this concept is described and claimed in U.S. Pat. No. 4,727,136, issued Feb. 23, 1988 to Jennings et al.). This vaccine also is immunogenic in sub-human primates, inducing serum antibodies that activate complement-mediated bacteriolysis (Fusco et al., 1997, J. Infect. Dis. 175: 364-372). In humans, such antibodies are known to confer protection against developing meningococcal disease (Goldschneider et al., 1969, J. Exp. Med. 129:1307). However, a subset of the antibodies induced by this vaccine have autoantibody activity to unmodified MenB PS (i.e. N-acetyl-MenB PS), Granoff et al., 1998, J. Immunol; 160: 5028-5036, which raise serious safety concerns about the use of this vaccine in humans. Therefore, investigators have sought alternative approaches to develop a safe and effective vaccine for prevention of disease caused by serogroup B strains.
Other groups have focused on surface proteins as vaccines. For example, the principal protein component of the pilus, pilin, elicits an immune response; however, so many antigenic variants exist and continue to develop that vaccines against the pilus protein have not been highly effective. See, U.S. Pat. No. 5,597,572. In other examples, vaccines have focused the highly conserved Neisserial surface protein A (NspA) (see, e.g., PCT Publication No. WO96/29412). Although the gene is highly conserved and expressed in virtually all strains, both polyclonal and monoclonal antibodies prepared against recombinant NspA are bactericidal and/or provide protection, against only about 50% of genetically diverse strains (Moe et al. (1999 Infect. Immun. 67: 5664; Moe et al. Infect Immun. 2001 69:3762). These observations suggest that recombinant NspA alone will not provide adequate protection against a broad spectrum of Neisserial strains.
Still other groups have used membrane preparations to induce immunity. In general, attempts to produce a meningococcal B vaccine based on outer membrane vesicles used repeated immunizations with material prepared from a single strain or repeated immunization with a vaccine containing vesicle antigens from multiple strains. When the vaccine contained vesicle antigens from more than strain, the resulting bactericidal antibody titers of infants or children given two or three doses were low (Cartwright K et al, 1999, Vaccine; 17:2612-2619; de Kleinjn E D et al, 2000, Vaccine, 18: 1456-1466), In these studies, and in a study done in cynomolgus monkeys (Rouupe van der Voort E R, 2000, Vaccine, 18:1334-1343) there also was evidence of immune interference between the responses to the different antigen. When repeated immunization with vesicles from a single strain was used, higher antibody titers resulted but the spectrum of antibody reactivity was limited to only a few strains that tended to be serologically similar to each other (Tappero et al., 1999, JAMA 281:1520; and Rouupe van der Voort E R, 2000, Vaccine, 18:1334-1343). Our experiments in laboratory animal models, which are described below confirmed this latter observation. Antisera from control animals given two sequential immunizations of a outer membrane vesicle vaccine prepared at the National Institute of Public Health, Oslo, Norway, from a single Neisseria meningitidis serogroup B strain, H44/76 (B:15:P1.7,16; “Norwegian vaccine”), reacted by flow cytometry and were bactericidal against only serogroup B strains that were of the same serosubtype (i.e. P1.7,16) or strains having an epitope similar to the P1.16 epitope (such as P1.10-4 strains).
Humans are the only known reservoir for Neisseria meningitidis spp. Accordingly, Neisserial species have evolved a wide variety of highly effective strategies to evade the human immune system. These include expression of a polysaccharide capsule that is cross-reactive with host polysialic acid (i.e. serogroup B) and high antigenic mutability for the immunodominant noncapsular epitopes, i.e. epitopes of antigens that are present at the surface in relatively large quantities, are accessible to antibodies, and elicit a strong antibody response.
Prior efforts to develop broad spectrum vaccines have been hampered by the wide variety of highly effective strategies used by Neisserial species to evade the human immune system. Because of these strategies, an immune response to a given strain will often not confer effective immunity against other strains of Neisseria. The present invention overcomes the disadvantages of prior art approaches to vaccination and elicits protective immunity against a broad spectrum of Neisseria meningitidis strains, notably (but not exclusively) including strains belonging to serogroup B.